July 2025 - Linux-stable-mirror

[PATCH stable] ipmi:msghandler: Fix potential memory corruption in ipmi_create_user()

by Brendan Jackman

From: Dan Carpenter <dan.carpenter(a)linaro.org> commit fa332f5dc6fc662ad7d3200048772c96b861cf6b upstream The "intf" list iterator is an invalid pointer if the correct "intf->intf_num" is not found. Calling atomic_dec(&intf->nr_users) on and invalid pointer will lead to memory corruption. We don't really need to call atomic_dec() if we haven't called atomic_add_return() so update the if (intf->in_shutdown) path as well. Fixes: 8e76741c3d8b ("ipmi: Add a limit on the number of users that may use IPMI") Signed-off-by: Dan Carpenter <dan.carpenter(a)linaro.org> Message-ID: <aBjMZ8RYrOt6NOgi(a)stanley.mountain> Signed-off-by: Corey Minyard <corey(a)minyard.net> Signed-off-by: Brendan Jackman <jackmanb(a)google.com> --- I have tested this in 6.12 with Google's platform drivers added to reproduce the bug. The bug causes the panic notifier chain to get corrupted leading to a crash. With the fix this goes away. Applies to 6.6 too but I haven't tested it there. Backport changes: - Dropped change to the `if (intf->in_shutdown)` block since that logic doesn't exist yet. - Modified out_unlock to release the srcu lock instead of the mutex since we don't have the mutex here yet. --- drivers/char/ipmi/ipmi_msghandler.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index e12b531f5c2f338008a42dc2c35b0a62395c9f3c..6a4a8ecd0edd02793eda70f9f9ae578e37da477f 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -1241,7 +1241,7 @@ int ipmi_create_user(unsigned int if_num, } /* Not found, return an error */ rv = -EINVAL; - goto out_kfree; + goto out_unlock; found: if (atomic_add_return(1, &intf->nr_users) > max_users) { @@ -1283,6 +1283,7 @@ int ipmi_create_user(unsigned int if_num, out_kfree: atomic_dec(&intf->nr_users); +out_unlock: srcu_read_unlock(&ipmi_interfaces_srcu, index); vfree(new_user); return rv; --- base-commit: 783cd2c3dca8b6c434e955b84c20c8940588dc68 change-id: 20250630-ipmi-fix-c565f7098afd Best regards, -- Brendan Jackman <jackmanb(a)google.com>

3 weeks, 1 day

3
5
0 0

[PATCH v2] drm/xe/bmg: fix compressed VRAM handling

by Matthew Auld

There looks to be an issue in our compression handling when the BO pages are very fragmented, where we choose to skip the identity map and instead fall back to emitting the PTEs by hand when migrating memory, such that we can hopefully do more work per blit operation. However in such a case we need to ensure the src PTEs are correctly tagged with a compression enabled PAT index on dgpu xe2+, otherwise the copy will simply treat the src memory as uncompressed, leading to corruption if the memory was compressed by the user. To fix this pass along use_comp_pat into emit_pte() on the src side, to indicate that compression should be considered. v2 (Jonathan): tweak the commit message Fixes: 523f191cc0c7 ("drm/xe/xe_migrate: Handle migration logic for xe2+ dgfx") Signed-off-by: Matthew Auld <matthew.auld(a)intel.com> Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray(a)intel.com> Cc: Thomas Hellström <thomas.hellstrom(a)linux.intel.com> Cc: Akshata Jahagirdar <akshata.jahagirdar(a)intel.com> Cc: <stable(a)vger.kernel.org> # v6.12+ Reviewed-by: Jonathan Cavitt <jonathan.cavitt(a)intel.com> --- drivers/gpu/drm/xe/xe_migrate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c index 0838582537e8..4e2bdf70eb70 100644 --- a/drivers/gpu/drm/xe/xe_migrate.c +++ b/drivers/gpu/drm/xe/xe_migrate.c @@ -863,7 +863,7 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m, if (src_is_vram && xe_migrate_allow_identity(src_L0, &src_it)) xe_res_next(&src_it, src_L0); else - emit_pte(m, bb, src_L0_pt, src_is_vram, copy_system_ccs, + emit_pte(m, bb, src_L0_pt, src_is_vram, copy_system_ccs || use_comp_pat, &src_it, src_L0, src); if (dst_is_vram && xe_migrate_allow_identity(src_L0, &dst_it)) -- 2.50.0

3 weeks, 1 day

1
0
0 0

[PATCH] x86/mm: Disable hugetlb page table sharing on non-PAE 32-bit

by Jann Horn

Only select ARCH_WANT_HUGE_PMD_SHARE if hugetlb page table sharing is actually possible; page table sharing requires at least three levels, because it involves shared references to PMD tables. Having ARCH_WANT_HUGE_PMD_SHARE enabled on non-PAE 32-bit X86 (which has 2-level paging) became particularly problematic after commit 59d9094df3d7 ("mm: hugetlb: independent PMD page table shared count"), since that changes `struct ptdesc` such that the `pt_mm` (for PGDs) and the `pt_share_count` (for PMDs) share the same union storage - and with 2-level paging, PMDs are PGDs. (For comparison, arm64 also gates ARCH_WANT_HUGE_PMD_SHARE on the configuration of page tables such that it is never enabled with 2-level paging.) Reported-by: Vitaly Chikunov <vt(a)altlinux.org> Closes: https://lore.kernel.org/r/srhpjxlqfna67blvma5frmy3aa@altlinux.org Fixes: cfe28c5d63d8 ("x86: mm: Remove x86 version of huge_pmd_share.") Cc: stable(a)vger.kernel.org Signed-off-by: Jann Horn <jannh(a)google.com> --- arch/x86/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 71019b3b54ea..917f523b994b 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -147,7 +147,7 @@ config X86 select ARCH_WANTS_DYNAMIC_TASK_STRUCT select ARCH_WANTS_NO_INSTR select ARCH_WANT_GENERAL_HUGETLB - select ARCH_WANT_HUGE_PMD_SHARE + select ARCH_WANT_HUGE_PMD_SHARE if PGTABLE_LEVELS > 2 select ARCH_WANT_LD_ORPHAN_WARN select ARCH_WANT_OPTIMIZE_DAX_VMEMMAP if X86_64 select ARCH_WANT_OPTIMIZE_HUGETLB_VMEMMAP if X86_64 --- base-commit: d0b3b7b22dfa1f4b515fd3a295b3fd958f9e81af change-id: 20250630-x86-2level-hugetlb-b1d8feb255ce -- Jann Horn <jannh(a)google.com>

3 weeks, 1 day

3
3
0 0

[PATCH] fs: Prevent file descriptor table allocations exceeding INT_MAX

by Sasha Levin

When sysctl_nr_open is set to a very high value (for example, 1073741816 as set by systemd), processes attempting to use file descriptors near the limit can trigger massive memory allocation attempts that exceed INT_MAX, resulting in a WARNING in mm/slub.c: WARNING: CPU: 0 PID: 44 at mm/slub.c:5027 __kvmalloc_node_noprof+0x21a/0x288 This happens because kvmalloc_array() and kvmalloc() check if the requested size exceeds INT_MAX and emit a warning when the allocation is not flagged with __GFP_NOWARN. Specifically, when nr_open is set to 1073741816 (0x3ffffff8) and a process calls dup2(oldfd, 1073741880), the kernel attempts to allocate: - File descriptor array: 1073741880 * 8 bytes = 8,589,935,040 bytes - Multiple bitmaps: ~400MB - Total allocation size: > 8GB (exceeding INT_MAX = 2,147,483,647) Reproducer: 1. Set /proc/sys/fs/nr_open to 1073741816: # echo 1073741816 > /proc/sys/fs/nr_open 2. Run a program that uses a high file descriptor: #include <unistd.h> #include <sys/resource.h> int main() { struct rlimit rlim = {1073741824, 1073741824}; setrlimit(RLIMIT_NOFILE, &rlim); dup2(2, 1073741880); // Triggers the warning return 0; } 3. Observe WARNING in dmesg at mm/slub.c:5027 systemd commit a8b627a introduced automatic bumping of fs.nr_open to the maximum possible value. The rationale was that systems with memory control groups (memcg) no longer need separate file descriptor limits since memory is properly accounted. However, this change overlooked that: 1. The kernel's allocation functions still enforce INT_MAX as a maximum size regardless of memcg accounting 2. Programs and tests that legitimately test file descriptor limits can inadvertently trigger massive allocations 3. The resulting allocations (>8GB) are impractical and will always fail systemd's algorithm starts with INT_MAX and keeps halving the value until the kernel accepts it. On most systems, this results in nr_open being set to 1073741816 (0x3ffffff8), which is just under 1GB of file descriptors. While processes rarely use file descriptors near this limit in normal operation, certain selftests (like tools/testing/selftests/core/unshare_test.c) and programs that test file descriptor limits can trigger this issue. Fix this by adding a check in alloc_fdtable() to ensure the requested allocation size does not exceed INT_MAX. This causes the operation to fail with -EMFILE instead of triggering a kernel warning and avoids the impractical >8GB memory allocation request. Fixes: 9cfe015aa424 ("get rid of NR_OPEN and introduce a sysctl_nr_open") Cc: stable(a)vger.kernel.org Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- fs/file.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/fs/file.c b/fs/file.c index b6db031545e65..6d2275c3be9c6 100644 --- a/fs/file.c +++ b/fs/file.c @@ -197,6 +197,21 @@ static struct fdtable *alloc_fdtable(unsigned int slots_wanted) return ERR_PTR(-EMFILE); } + /* + * Check if the allocation size would exceed INT_MAX. kvmalloc_array() + * and kvmalloc() will warn if the allocation size is greater than + * INT_MAX, as filp_cachep objects are not __GFP_NOWARN. + * + * This can happen when sysctl_nr_open is set to a very high value and + * a process tries to use a file descriptor near that limit. For example, + * if sysctl_nr_open is set to 1073741816 (0x3ffffff8) - which is what + * systemd typically sets it to - then trying to use a file descriptor + * close to that value will require allocating a file descriptor table + * that exceeds 8GB in size. + */ + if (unlikely(nr > INT_MAX / sizeof(struct file *))) + return ERR_PTR(-EMFILE); + fdt = kmalloc(sizeof(struct fdtable), GFP_KERNEL_ACCOUNT); if (!fdt) goto out; -- 2.39.5

3 weeks, 1 day

3
5
0 0

[PATCH 1/1] usb: cdnsp: do not disable slot for disabled slot

by Peter Chen

It doesn't need to do it, and the related command event returns 'Slot Not Enabled Error' status. Fixes: 3d82904559f4 ("usb: cdnsp: cdns3 Add main part of Cadence USBSSP DRD Driver") cc: stable(a)vger.kernel.org Suggested-by: Hongliang Yang <hongliang.yang(a)cixtech.com> Signed-off-by: Peter Chen <peter.chen(a)cixtech.com> --- drivers/usb/cdns3/cdnsp-ring.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/usb/cdns3/cdnsp-ring.c b/drivers/usb/cdns3/cdnsp-ring.c index fd06cb85c4ea..757fdd918286 100644 --- a/drivers/usb/cdns3/cdnsp-ring.c +++ b/drivers/usb/cdns3/cdnsp-ring.c @@ -772,7 +772,9 @@ static int cdnsp_update_port_id(struct cdnsp_device *pdev, u32 port_id) } if (port_id != old_port) { - cdnsp_disable_slot(pdev); + if (pdev->slot_id) + cdnsp_disable_slot(pdev); + pdev->active_port = port; cdnsp_enable_slot(pdev); } -- 2.25.1

3 weeks, 1 day

3
3
0 0

[PATCH] net: libwx: fix the incorrect display of the queue number

by Jiawen Wu

When setting "ethtool -L eth0 combined 1", the number of RX/TX queue is changed to be 1. RSS is disabled at this moment, and the indices of FDIR have not be changed in wx_set_rss_queues(). So the combined count still shows the previous value. This issue was introduced when supporting FDIR. Fix it for those devices that support FDIR. Fixes: 34744a7749b3 ("net: txgbe: add FDIR info to ethtool ops") Cc: stable(a)vger.kernel.org Signed-off-by: Jiawen Wu <jiawenwu(a)trustnetic.com> --- drivers/net/ethernet/wangxun/libwx/wx_ethtool.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c b/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c index c12a4cb951f6..d9de600e685a 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c @@ -438,6 +438,10 @@ void wx_get_channels(struct net_device *dev, /* record RSS queues */ ch->combined_count = wx->ring_feature[RING_F_RSS].indices; + /* nothing else to report if RSS is disabled */ + if (ch->combined_count == 1) + return; + if (test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags)) ch->combined_count = wx->ring_feature[RING_F_FDIR].indices; } -- 2.48.1

3 weeks, 1 day

3
3
0 0

[PATCH net v4 2/3] net: wangxun: revert the adjustment of the IRQ vector sequence

by Jiawen Wu

Due to hardware limitations of NGBE, queue IRQs can only be requested on vector 0 to 7. When the number of queues is set to the maximum 8, the PCI IRQ vectors are allocated from 0 to 8. The vector 0 is used by MISC interrupt, and althrough the vector 8 is used by queue interrupt, it is unable to receive packets. This will cause some packets to be dropped when RSS is enabled and they are assigned to queue 8. So revert the adjustment of the MISC IRQ location, to make it be the last one in IRQ vectors. Fixes: 937d46ecc5f9 ("net: wangxun: add ethtool_ops for channel number") Cc: stable(a)vger.kernel.org Signed-off-by: Jiawen Wu <jiawenwu(a)trustnetic.com> Reviewed-by: Larysa Zaremba <larysa.zaremba(a)intel.com> --- drivers/net/ethernet/wangxun/libwx/wx_lib.c | 17 ++++++++--------- drivers/net/ethernet/wangxun/libwx/wx_type.h | 2 +- drivers/net/ethernet/wangxun/ngbe/ngbe_main.c | 2 +- drivers/net/ethernet/wangxun/ngbe/ngbe_type.h | 2 +- drivers/net/ethernet/wangxun/txgbe/txgbe_irq.c | 6 +++--- drivers/net/ethernet/wangxun/txgbe/txgbe_type.h | 4 ++-- 6 files changed, 16 insertions(+), 17 deletions(-) diff --git a/drivers/net/ethernet/wangxun/libwx/wx_lib.c b/drivers/net/ethernet/wangxun/libwx/wx_lib.c index 7f2e6cddfeb1..66eaf5446115 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_lib.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_lib.c @@ -1746,7 +1746,7 @@ static void wx_set_num_queues(struct wx *wx) */ static int wx_acquire_msix_vectors(struct wx *wx) { - struct irq_affinity affd = { .pre_vectors = 1 }; + struct irq_affinity affd = { .post_vectors = 1 }; int nvecs, i; /* We start by asking for one vector per queue pair */ @@ -1783,16 +1783,17 @@ static int wx_acquire_msix_vectors(struct wx *wx) return nvecs; } - wx->msix_entry->entry = 0; - wx->msix_entry->vector = pci_irq_vector(wx->pdev, 0); nvecs -= 1; for (i = 0; i < nvecs; i++) { wx->msix_q_entries[i].entry = i; - wx->msix_q_entries[i].vector = pci_irq_vector(wx->pdev, i + 1); + wx->msix_q_entries[i].vector = pci_irq_vector(wx->pdev, i); } wx->num_q_vectors = nvecs; + wx->msix_entry->entry = nvecs; + wx->msix_entry->vector = pci_irq_vector(wx->pdev, nvecs); + return 0; } @@ -2299,8 +2300,6 @@ static void wx_set_ivar(struct wx *wx, s8 direction, wr32(wx, WX_PX_MISC_IVAR, ivar); } else { /* tx or rx causes */ - if (!(wx->mac.type == wx_mac_em && wx->num_vfs == 7)) - msix_vector += 1; /* offset for queue vectors */ msix_vector |= WX_PX_IVAR_ALLOC_VAL; index = ((16 * (queue & 1)) + (8 * direction)); ivar = rd32(wx, WX_PX_IVAR(queue >> 1)); @@ -2339,7 +2338,7 @@ void wx_write_eitr(struct wx_q_vector *q_vector) itr_reg |= WX_PX_ITR_CNT_WDIS; - wr32(wx, WX_PX_ITR(v_idx + 1), itr_reg); + wr32(wx, WX_PX_ITR(v_idx), itr_reg); } /** @@ -2392,9 +2391,9 @@ void wx_configure_vectors(struct wx *wx) wx_write_eitr(q_vector); } - wx_set_ivar(wx, -1, 0, 0); + wx_set_ivar(wx, -1, 0, v_idx); if (pdev->msix_enabled) - wr32(wx, WX_PX_ITR(0), 1950); + wr32(wx, WX_PX_ITR(v_idx), 1950); } EXPORT_SYMBOL(wx_configure_vectors); diff --git a/drivers/net/ethernet/wangxun/libwx/wx_type.h b/drivers/net/ethernet/wangxun/libwx/wx_type.h index 7730c9fc3e02..d392394791b3 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_type.h +++ b/drivers/net/ethernet/wangxun/libwx/wx_type.h @@ -1343,7 +1343,7 @@ struct wx { }; #define WX_INTR_ALL (~0ULL) -#define WX_INTR_Q(i) BIT((i) + 1) +#define WX_INTR_Q(i) BIT((i)) /* register operations */ #define wr32(a, reg, value) writel((value), ((a)->hw_addr + (reg))) diff --git a/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c b/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c index b5022c49dc5e..68415a7ef12f 100644 --- a/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c +++ b/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c @@ -161,7 +161,7 @@ static void ngbe_irq_enable(struct wx *wx, bool queues) if (queues) wx_intr_enable(wx, NGBE_INTR_ALL); else - wx_intr_enable(wx, NGBE_INTR_MISC); + wx_intr_enable(wx, NGBE_INTR_MISC(wx)); } /** diff --git a/drivers/net/ethernet/wangxun/ngbe/ngbe_type.h b/drivers/net/ethernet/wangxun/ngbe/ngbe_type.h index bb74263f0498..6eca6de475f7 100644 --- a/drivers/net/ethernet/wangxun/ngbe/ngbe_type.h +++ b/drivers/net/ethernet/wangxun/ngbe/ngbe_type.h @@ -87,7 +87,7 @@ #define NGBE_PX_MISC_IC_TIMESYNC BIT(11) /* time sync */ #define NGBE_INTR_ALL 0x1FF -#define NGBE_INTR_MISC BIT(0) +#define NGBE_INTR_MISC(A) BIT((A)->num_q_vectors) #define NGBE_PHY_CONFIG(reg_offset) (0x14000 + ((reg_offset) * 4)) #define NGBE_CFG_LAN_SPEED 0x14440 diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_irq.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_irq.c index dc468053bdf8..3885283681ec 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_irq.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_irq.c @@ -31,7 +31,7 @@ void txgbe_irq_enable(struct wx *wx, bool queues) wr32(wx, WX_PX_MISC_IEN, misc_ien); /* unmask interrupt */ - wx_intr_enable(wx, TXGBE_INTR_MISC); + wx_intr_enable(wx, TXGBE_INTR_MISC(wx)); if (queues) wx_intr_enable(wx, TXGBE_INTR_QALL(wx)); } @@ -131,7 +131,7 @@ static irqreturn_t txgbe_misc_irq_handle(int irq, void *data) txgbe->eicr = eicr; if (eicr & TXGBE_PX_MISC_IC_VF_MBOX) { wx_msg_task(txgbe->wx); - wx_intr_enable(wx, TXGBE_INTR_MISC); + wx_intr_enable(wx, TXGBE_INTR_MISC(wx)); } return IRQ_WAKE_THREAD; } @@ -183,7 +183,7 @@ static irqreturn_t txgbe_misc_irq_thread_fn(int irq, void *data) nhandled++; } - wx_intr_enable(wx, TXGBE_INTR_MISC); + wx_intr_enable(wx, TXGBE_INTR_MISC(wx)); return (nhandled > 0 ? IRQ_HANDLED : IRQ_NONE); } diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h index 42ec815159e8..41915d7dd372 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h @@ -302,8 +302,8 @@ struct txgbe_fdir_filter { #define TXGBE_DEFAULT_RX_WORK 128 #endif -#define TXGBE_INTR_MISC BIT(0) -#define TXGBE_INTR_QALL(A) GENMASK((A)->num_q_vectors, 1) +#define TXGBE_INTR_MISC(A) BIT((A)->num_q_vectors) +#define TXGBE_INTR_QALL(A) (TXGBE_INTR_MISC(A) - 1) #define TXGBE_MAX_EITR GENMASK(11, 3) -- 2.48.1

3 weeks, 1 day

1
0
0 0

[PATCH net v4 1/3] net: txgbe: request MISC IRQ in ndo_open

by Jiawen Wu

Move the creating of irq_domain for MISC IRQ from .probe to .ndo_open, and free it in .ndo_stop, to maintain consistency with the queue IRQs. This it for subsequent adjustments to the IRQ vectors. Fixes: aefd013624a1 ("net: txgbe: use irq_domain for interrupt controller") Cc: stable(a)vger.kernel.org Signed-off-by: Jiawen Wu <jiawenwu(a)trustnetic.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski(a)linux.intel.com> --- .../net/ethernet/wangxun/txgbe/txgbe_irq.c | 2 +- .../net/ethernet/wangxun/txgbe/txgbe_main.c | 22 +++++++++---------- 2 files changed, 11 insertions(+), 13 deletions(-) diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_irq.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_irq.c index 20b9a28bcb55..dc468053bdf8 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_irq.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_irq.c @@ -78,7 +78,6 @@ int txgbe_request_queue_irqs(struct wx *wx) free_irq(wx->msix_q_entries[vector].vector, wx->q_vector[vector]); } - wx_reset_interrupt_capability(wx); return err; } @@ -211,6 +210,7 @@ void txgbe_free_misc_irq(struct txgbe *txgbe) free_irq(txgbe->link_irq, txgbe); free_irq(txgbe->misc.irq, txgbe); txgbe_del_irq_domain(txgbe); + txgbe->wx->misc_irq_domain = false; } int txgbe_setup_misc_irq(struct txgbe *txgbe) diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c index f3d2778b8e35..a5867f3c93fc 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c @@ -458,10 +458,14 @@ static int txgbe_open(struct net_device *netdev) wx_configure(wx); - err = txgbe_request_queue_irqs(wx); + err = txgbe_setup_misc_irq(wx->priv); if (err) goto err_free_resources; + err = txgbe_request_queue_irqs(wx); + if (err) + goto err_free_misc_irq; + /* Notify the stack of the actual queue counts. */ err = netif_set_real_num_tx_queues(netdev, wx->num_tx_queues); if (err) @@ -479,6 +483,9 @@ static int txgbe_open(struct net_device *netdev) err_free_irq: wx_free_irq(wx); +err_free_misc_irq: + txgbe_free_misc_irq(wx->priv); + wx_reset_interrupt_capability(wx); err_free_resources: wx_free_resources(wx); err_reset: @@ -519,6 +526,7 @@ static int txgbe_close(struct net_device *netdev) wx_ptp_stop(wx); txgbe_down(wx); wx_free_irq(wx); + txgbe_free_misc_irq(wx->priv); wx_free_resources(wx); txgbe_fdir_filter_exit(wx); wx_control_hw(wx, false); @@ -564,7 +572,6 @@ static void txgbe_shutdown(struct pci_dev *pdev) int txgbe_setup_tc(struct net_device *dev, u8 tc) { struct wx *wx = netdev_priv(dev); - struct txgbe *txgbe = wx->priv; /* Hardware has to reinitialize queues and interrupts to * match packet buffer alignment. Unfortunately, the @@ -575,7 +582,6 @@ int txgbe_setup_tc(struct net_device *dev, u8 tc) else txgbe_reset(wx); - txgbe_free_misc_irq(txgbe); wx_clear_interrupt_scheme(wx); if (tc) @@ -584,7 +590,6 @@ int txgbe_setup_tc(struct net_device *dev, u8 tc) netdev_reset_tc(dev); wx_init_interrupt_scheme(wx); - txgbe_setup_misc_irq(txgbe); if (netif_running(dev)) txgbe_open(dev); @@ -882,13 +887,9 @@ static int txgbe_probe(struct pci_dev *pdev, txgbe_init_fdir(txgbe); - err = txgbe_setup_misc_irq(txgbe); - if (err) - goto err_release_hw; - err = txgbe_init_phy(txgbe); if (err) - goto err_free_misc_irq; + goto err_release_hw; err = register_netdev(netdev); if (err) @@ -916,8 +917,6 @@ static int txgbe_probe(struct pci_dev *pdev, err_remove_phy: txgbe_remove_phy(txgbe); -err_free_misc_irq: - txgbe_free_misc_irq(txgbe); err_release_hw: wx_clear_interrupt_scheme(wx); wx_control_hw(wx, false); @@ -957,7 +956,6 @@ static void txgbe_remove(struct pci_dev *pdev) unregister_netdev(netdev); txgbe_remove_phy(txgbe); - txgbe_free_misc_irq(txgbe); wx_free_isb_resources(wx); pci_release_selected_regions(pdev, -- 2.48.1

3 weeks, 1 day

1
0
0 0

[PATCH v2] block: Fix a deadlock related to modifying the readahead attribute

by Bart Van Assche

Every time I run test srp/002 the following deadlock is triggered: task:multipathd Call Trace: <TASK> __schedule+0x8c1/0x1bf0 schedule+0xdd/0x270 schedule_preempt_disabled+0x1c/0x30 __mutex_lock+0xb89/0x1650 mutex_lock_nested+0x1f/0x30 dm_table_set_restrictions+0x823/0xdf0 __bind+0x166/0x590 dm_swap_table+0x2a7/0x490 do_resume+0x1b1/0x610 dev_suspend+0x55/0x1a0 ctl_ioctl+0x3a5/0x7e0 dm_ctl_ioctl+0x12/0x20 __x64_sys_ioctl+0x127/0x1a0 x64_sys_call+0xe2b/0x17d0 do_syscall_64+0x96/0x3a0 entry_SYSCALL_64_after_hwframe+0x4b/0x53 </TASK> task:(udev-worker) Call Trace: <TASK> __schedule+0x8c1/0x1bf0 schedule+0xdd/0x270 blk_mq_freeze_queue_wait+0xf2/0x140 blk_mq_freeze_queue_nomemsave+0x23/0x30 queue_ra_store+0x14e/0x290 queue_attr_store+0x23e/0x2c0 sysfs_kf_write+0xde/0x140 kernfs_fop_write_iter+0x3b2/0x630 vfs_write+0x4fd/0x1390 ksys_write+0xfd/0x230 __x64_sys_write+0x76/0xc0 x64_sys_call+0x276/0x17d0 do_syscall_64+0x96/0x3a0 entry_SYSCALL_64_after_hwframe+0x4b/0x53 </TASK> This deadlock happens because blk_mq_freeze_queue_nomemsave() waits for pending requests to finish. The pending requests do never complete because the dm-multipath queue_if_no_path option is enabled and the only path in the dm-multipath configuration is being removed. Fix this deadlock by removing the queue freezing/unfreezing code from queue_ra_store(). Freezing the request queue from inside a block layer sysfs store callback function is essential when modifying parameters that affect how bios or requests are processed, e.g. parameters that affect bio_split_to_limit(). Freezing the request queue when modifying parameters that do not affect bio nor request processing is not necessary. Cc: Nilay Shroff <nilay(a)linux.ibm.com> Cc: stable(a)vger.kernel.org Fixes: b07a889e8335 ("block: move q->sysfs_lock and queue-freeze under show/store method") Signed-off-by: Bart Van Assche <bvanassche(a)acm.org> --- Changes compared to v1: made the patch description more detailed. block/blk-sysfs.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index b2b9b89d6967..1f63b184c6e9 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -105,7 +105,6 @@ queue_ra_store(struct gendisk *disk, const char *page, size_t count) { unsigned long ra_kb; ssize_t ret; - unsigned int memflags; struct request_queue *q = disk->queue; ret = queue_var_store(&ra_kb, page, count); @@ -116,10 +115,8 @@ queue_ra_store(struct gendisk *disk, const char *page, size_t count) * calculated from the queue limits by queue_limits_commit_update. */ mutex_lock(&q->limits_lock); - memflags = blk_mq_freeze_queue(q); disk->bdi->ra_pages = ra_kb >> (PAGE_SHIFT - 10); mutex_unlock(&q->limits_lock); - blk_mq_unfreeze_queue(q, memflags); return ret; }

3 weeks, 1 day

4
7
0 0

[PATCH] block: Fix a deadlock related to modifying the readahead attribute

by Bart Van Assche

Every time I run test srp/002 the following deadlock is triggered: task:multipathd Call Trace: <TASK> __schedule+0x8c1/0x1bf0 schedule+0xdd/0x270 schedule_preempt_disabled+0x1c/0x30 __mutex_lock+0xb89/0x1650 mutex_lock_nested+0x1f/0x30 dm_table_set_restrictions+0x823/0xdf0 __bind+0x166/0x590 dm_swap_table+0x2a7/0x490 do_resume+0x1b1/0x610 dev_suspend+0x55/0x1a0 ctl_ioctl+0x3a5/0x7e0 dm_ctl_ioctl+0x12/0x20 __x64_sys_ioctl+0x127/0x1a0 x64_sys_call+0xe2b/0x17d0 do_syscall_64+0x96/0x3a0 entry_SYSCALL_64_after_hwframe+0x4b/0x53 </TASK> task:(udev-worker) Call Trace: <TASK> __schedule+0x8c1/0x1bf0 schedule+0xdd/0x270 blk_mq_freeze_queue_wait+0xf2/0x140 blk_mq_freeze_queue_nomemsave+0x23/0x30 queue_ra_store+0x14e/0x290 queue_attr_store+0x23e/0x2c0 sysfs_kf_write+0xde/0x140 kernfs_fop_write_iter+0x3b2/0x630 vfs_write+0x4fd/0x1390 ksys_write+0xfd/0x230 __x64_sys_write+0x76/0xc0 x64_sys_call+0x276/0x17d0 do_syscall_64+0x96/0x3a0 entry_SYSCALL_64_after_hwframe+0x4b/0x53 </TASK> Fix this by removing the superfluous queue freezing/unfreezing code from queue_ra_store(). Cc: Nilay Shroff <nilay(a)linux.ibm.com> Cc: stable(a)vger.kernel.org Fixes: b07a889e8335 ("block: move q->sysfs_lock and queue-freeze under show/store method") Signed-off-by: Bart Van Assche <bvanassche(a)acm.org> --- block/blk-sysfs.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index b2b9b89d6967..1f63b184c6e9 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -105,7 +105,6 @@ queue_ra_store(struct gendisk *disk, const char *page, size_t count) { unsigned long ra_kb; ssize_t ret; - unsigned int memflags; struct request_queue *q = disk->queue; ret = queue_var_store(&ra_kb, page, count); @@ -116,10 +115,8 @@ queue_ra_store(struct gendisk *disk, const char *page, size_t count) * calculated from the queue limits by queue_limits_commit_update. */ mutex_lock(&q->limits_lock); - memflags = blk_mq_freeze_queue(q); disk->bdi->ra_pages = ra_kb >> (PAGE_SHIFT - 10); mutex_unlock(&q->limits_lock); - blk_mq_unfreeze_queue(q, memflags); return ret; }

3 weeks, 1 day

3
9
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror July 2025